Quantum Physics
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Monday 31 August
Practical robust self-testing of multipartite entanglement has so far been restricted to small-scale systems due to error bounds that degrade severely with system size. In this work, we establish multipartite self-testing with robustness independent of the size of the quantum network. We derive a fully analytic, device-independent self-testing bound for $n$-qubit Greenberger-Horne-Zeilinger (GHZ) states. The bound scales linearly with the observed violation error and lies universally within a constant factor of two from a theoretical upper bound. Furthermore, the operator-inequality framework reduces the verification of the conjectured optimal bound to a highly efficient numerical check, which we perform up to $n=100$. Consequently, GHZ entanglement can be certified under a fixed noise level in arbitrarily large systems, enabling scalable device-independent verification.
High Energy Physics (Phenomenology)
Exclusive Leptonium Electroproduction
Purely leptonic bound states provide precision probes of QED. Positronium $(e^+e^-)$ and muonium $(\mu^+e^-)$ have long been observed, whereas dimuonium $(\mu^+\mu^-)$ and tauonium $(\tau^+\tau^-)$ remain undiscovered. We study exclusive vector-leptonium electroproduction in $ep$ collisions within nonrelativistic QED. We include the Bethe–Heitler and double deeply virtual Compton scattering contributions and their interference, and calculate the NLO QCD hard-scattering kernels entering the dominant Compton form factor $\Hcal$ within collinear GPD factorization. The NLO QCD correction to the DDVCS contribution changes from a strong suppression at low photon virtuality to a sizable enhancement as the lower virtuality cut is raised, with the gluon channel providing the dominant contribution. Bethe–Heitler production dominates the exclusive rate, supporting dedicated dimuonium searches at the EIC and JLab, with larger samples expected at higher-energy electron–proton colliders. The much larger positronium samples provide a high-statistics environment for precision QED studies, whereas tauonium production remains strongly suppressed.
Read the full abstract
Statistical Mechanics
Inertia-Driven Information Flow and Symmetry Breaking in a Nonequilibrium Two-Bead System
We investigate information-flow generation in a nonequilibrium two-bead system coupled to two heat baths. We show that the system acts as an information-flow generator in both overdamped and underdamped regimes, with the underdamped dynamics revealing a divergence of the scaled information flow along specific paths in the thermal asymmetry–inertia parameter space that is hidden in the overdamped limit. This information generation suggests a possible route toward information engines and demon-like mechanisms in nanomachines. A symmetry-perturbation analysis of the response landscape of information flow reveals a geometric structure reminiscent of a Ginzburg–Landau framework: the symmetric reference state can correspond to a minimum or maximum depending on the perturbation direction, while the flat overdamped landscape develops a finite curvature under inertia. Mass asymmetry shifts the resulting maxima, and Hessian eigenvalue and eigenvector analysis reveals level touching of principal modes and bimodality along a constant-diffusion path. These results establish a minimal framework for understanding how inertia and microscopic heterogeneity shape information landscapes in nonequilibrium systems, with potential extensions to more complex heterogeneous networks.
Read the full abstract
Probability
Limit Laws of the Iterated Logarithm Under Sub-linear Expectations
Let $\{Y_n; n\ge 1\}$ be a sequence of independent and identically distributed random variables with mean zero in Peng's framework of the sub-linear expectation space $(\Omega,\mathscr{H},\widehat{\mathbb E})$, and $S_n=\sum_{i=1}^nY_i$. In this paper, we establish a limit law of \begin{align*}\lim_{n\to \infty}\max_{k\le n}\frac{S_k}{\sqrt{2k \log\log n}}. \end{align*} Different from the result obtained by Chen (2015) in which the limit is a constant, it is shown that under the upper capacity the limit may be prescribed as a given function of $Y_1,Y_2,\ldots$, taking values in the standard deviation interval. As a result, it is also shown that the set of limit points in the compact law of the iterated logarithm can be a symmetric random interval. This paper (Chinese version) has been submitted to Special Issue of Science in China-Mathematics in Celebration of Professor Peng Shige's 80th Birthday. In Theorem 2.2 of the original paper, an additional condition (2.6) is needed.
Read the full abstract
Analysis of PDEs
An Almgren-type formula for planar $p$-harmonic functions
For a nonconstant planar $p$-harmonic function $u$, with $1 < p < \infty$, and according to previous results, most notably Aronsson's fundamental analysis of the hodograph representation in a neighborhood of an isolated critical point, we introduce the flux-normalized frequency $$ N_*(x_0,r) = \frac{r\displaystyle\int_{B_r(x_0)} |D u|^p\,dx} {\displaystyle\int_{\partial B_r(x_0)} |D u|^{p-2}(u-u(x_0))^2\,dS}. $$ This ratio constitutes the natural counterpart of Almgren's frequency function in the nonlinear setting, as it recovers precisely the degree of homogeneity for every homogeneous $p$-harmonic profile. We further establish the corresponding gauge-corrected version of Almgren-type monotonicity and reinterpret, in terms of the frequency function, the classical planar unique continuation property. In addition, we derive a quantitative pinching estimate for the flux-normalized frequency and identify the specific obstruction that prevents a straightforward extension of the method to higher dimensions.
Read the full abstract
Artificial Intelligence
VFR-Audit: Verdict-Level Reliability for Fairness Audits in Hospital Length-of-Stay Prediction
Fairness audits in clinical Artificial Intelligence convert continuous fairness metrics into binary pass-or-fail verdicts against operational thresholds, where hospital governance boards, payers, and regulators act on the resulting verdicts. Such audits are repeated over time and across hospital sites, thus the same verdict can flip between pass and fail across audits. Existing uncertainty methods such as Bayesian posteriors, bootstrap confidence intervals, and permutation tests address verdict instability only at the continuous-metric level. Converting metric-level uncertainty into a verdict-stability claim remains a manual step that scales poorly across the (model, metric, attribute) cells an audit covers. Existing uncertainty methods also leave open whether bias-mitigation steps, such as reweighing or per-group threshold shifts, yield a stable passing verdict at the cost of model discrimination measured as AUROC or AUPRC.To address this verdict-stability gap, we propose VFR-Audit, a framework built around the Verdict Flip Rate (VFR), a scalar bounded between 0 and 0.5 that measures the probability of verdict reversal under stratified bootstrap resampling. VFR-Audit reports VFR alongside three reliability axes, namely within-cohort resampling stability, audit-size sensitivity, and cross-hospital verdict agreement via Fleiss' kappa.
Read the full abstract
Statistical Computation
Scalable Statistical Inference in Stochastic Gradient Descent
Constructing confidence regions for stochastic gradient descent (SGD) ideally requires estimating the asymptotic covariance matrix, a severe computational bottleneck in high dimensions. Traditional cancellation-based batch means methods bypass this estimation but require inverting a sample batch covariance matrix. This introduces strict mathematical degeneracy when the parameter dimension exceeds the number of batches. To address this problem, we utilize equal batch size batch means method and propose a simultaneous, marginal-friendly framework. The proposed marginal statistics has a asymptotic Student's $t$-distribution, and eliminates the matrix inversion step, entirely circumventing high-dimensional degeneracy. To achieve valid simultaneous coverage, we present an algorithm utilizing wild bootstrap samples drawn from a statistic as a function of only the diagonals of the variance-covariance estimator, and to further incorporate the contribution of cross-dependencies, we introduce an efficient Quasi-Monte Carlo procedure utilizing a $t$-copula approximation. Additionally, we integrate a Lugsail variance estimator to aggressively correct finite-sample bias and under-coverage. The proposed methodology delivers interpretable, simultaneous hyper-rectangular confidence regions that are statistically robust, memory-efficient, and strictly scalable for high-dimensional inference. The theoretical results are supported by extensive numerical simulation analysis through various aspects of dimension, number of batches and error structure.
Read the full abstract
Computer Vision
Pretrained, Curriculum-Tuned, and Ensembled: A Tracer-Aware Interactive Segmentation Pipeline for AutoPET V
Interactive lesion segmentation in whole-body PET/CT requires a model to provide a strong initial prediction while also responding efficiently to sparse corrective scribbles during inference. This setting is particularly challenging because tracer distributions, physiological uptake patterns, lesion appearance, and acquisition characteristics differ substantially between FDG and PSMA studies. We present TRIAGE, Tracer-aware Refinement via Interactive Anatomy-Guided sEgmentation. The core backbone is a 3D STU-Net initialized through masked autoencoding pre-training with an asynchronous masking strategy, aiming to learn transferable anatomical and cross-modal representations before task-specific fine-tuning. In parallel, we train an auxiliary organ segmentation model whose predictions provide explicit anatomical context and help distinguish physiological uptake from malignant lesions. A dedicated tracer classifier first routes each study to an FDG- or PSMA-specific branch. Within each branch, a first-stage segmentation model consumes CT, PET, and organ context to generate an initial lesion mask. The initial prediction is then combined with cumulative foreground/background scribbles and refined by a second interactive segmentation network. The FDG and PSMA branches share the same overall processing pipeline but are trained independently to account for tracer-specific appearance and error modes. We additionally employ curriculum-style training and model ensembling to improve robustness across interaction steps and heterogeneous cohorts. Experiments are conducted using the official AutoPET V data and ten-fold split; quantitative results, ablations, and final test-set performance are left as placeholders to be completed after the challenge evaluation. Code: https://github.com/Liiiii2101/AUTOPET2026-MEDAI.
Read the full abstract
Optimization and Control
On the real angular stability radius for sectorial matrices
The real stability radius problem asks for the smallest real perturbation that makes a given matrix lose rank, and it is a structured counterpart of the corresponding complex stability radius problem. In this paper, we consider an angular version of this question for sectorial matrices, where the size of a perturbation is measured by its largest phase instead of its largest singular value. More precisely, for a sectorial matrix $M$, we investigate the minimum required largest phase of a real and sectorial perturbation $\Delta$ for which $I+M\Delta$ loses rank. We show that for some matrices, the complex and the real angular stability radius are the same, but that they are in general different, such as when $M$ is diagonal or complex symmetric. For these cases, we derive a formula for the real angular stability radius, and this formula depends on the two largest phases of $M$. Moreover, in these cases we show that the result for $n \times n$ matrices can be reduced to that for $2 \times 2$ matrices by compressing an arbitrary real destabilizing perturbation to the real two-dimensional subspace spanned by the real and imaginary parts of a destabilizing vector.
Read the full abstract
Computation and Language
Thesis Proposal: Toward a Human-Centered and Perspective-Aware Framework for Reproducible ML Evaluation and AI Alignment
Humans play a vital role at every stage of AI development, from data collection and curation to model development and evaluation. However, humans often disagree with each other and sometimes with themselves over time. It is essential to take disagreement into account when building human-centered AI systems, especially in domains where it is prevalent, such as AI safety, content moderation, or sentiment analysis. Disagreement often arises from subjective human opinion and can vary with one's identity, beliefs, and social environment. Despite this, current LLM evaluation approaches frequently rely on aggregating labels (often via plurality voting) to represent consensus, thereby obscuring minority perspectives. By failing to account for human disagreement, these evaluation methods contribute to the reproducibility crisis in AI. Human feedback is also crucial for ensuring that AI systems align with human values. For these systems to be trustworthy, it is critical to ensure that they reflect diverse human values and perspectives. In this thesis proposal, we present a human-centered and perspective-aware framework for reproducible ML evaluation and AI alignment.
Read the full abstract
Artificial Intelligence
HSRM: Hidden-State Reward Models for Test-Time Verification
Large language models can often generate plausible mathematical reasoning traces, but reliably identifying the correct solution among multiple candidates remains a key challenge. Existing test-time reasoning pipelines typically rely on text-based verifiers that re-read each generated solution, making verification an expensive component of inference. Prior work has shown, however, that LLMs often encode correctness-related signals in their internal representations, including awareness of when their own answers are likely to be wrong. Building on this observation, we introduce HSRM, a lightweight hidden-state reward model that verifies candidate solutions by directly reading the generator's internal representations rather than re-processing its text. HSRM extracts hidden states from a frozen generator at reasoning-step boundaries and uses a small Transformer encoder to rank candidates. It is trained from self-generated trajectories with outcome labels, requiring neither human-written process supervision nor a large pretrained verifier. Across four mathematical reasoning benchmarks, HSRM matches or outperforms a 55M-parameter text-only energy verifier in 15 of 16 generator–dataset settings while using only about 2M parameters, providing an efficient alternative to text-only verification by reusing representations already computed during generation.
Read the full abstract
Fluid Dynamics
Real Schur Flow Acoustic Directivity and Helicity Effects within
The statistical description and acoustic diagnostics of anisotropic turbulence remain challenging. Real Schur flows (RSFs), prototype models of component-wise dimensionally reduced flows (CWDRFs), provide effective platforms for studying such problems. We combine absolute statistical equilibrium with the acoustic analogy to investigate helicity effects on the far-field directivity of sound radiated from three types of CWDRFs: two different RSFs, and a further reduction of them (liberated Schur flow). We evaluate the directivity from the vortex-sound source spectrum which is calculated with the Wick theorem from the absolute equilibirum Gaussian ensemble. Helicity amplifies the dimensional reduction induced anisotropy of the directivity functions across all CWDRFs, with distinct patterns.
Read the full abstract
Computer Vision
Physical Adversarial Examples for Person Detectors in Thermal Images Based on 3D Modeling
Thermal Infrared detection is widely used in autonomous driving, medical AI, etc., but its security has only attracted attention recently. We propose infrared adversarial clothing designed to evade thermal person detectors in real-world scenarios. The design of the adversarial clothing is based on 3D modeling, which makes it easier to simulate multiangle scenes near the real world compared to 2D modeling. We optimized the black patch layout pattern of 3D clothing based on the adversarial example technique and made physical adversarial clothing using the aerogel. The idea is to paste a set of square aerogel patches, which display black squares in thermal images, in the inner side of clothing at specific locations with specific orientations. To enhance realism, we propose a method to build infrared 3D models with real infrared photos and develop texture maps for 3D models to simulate varied infrared characteristics over time and location. In physical attacks, we achieved an attack success rate of 80.11% indoors and 76.85% outdoors against YOLOv9. In contrast, randomly placed patches yielded much lower success rates (26.53% indoors and 23.03% outdoors). The adversarial clothing also showed good transferability to unknown detectors with an ensemble attack method, demonstrating the effectiveness of our approach.
Read the full abstract
Cosmology and Nongalactic Astrophysics
Cosmic variance and ergodicity in finite systems with correlations
We consider the difference between ensemble and volume average in cosmology. It is known that for sufficiently weak long-range correlations the root mean square of the difference, which we call ergodicity bias, decays like $R^{-3/2}$ in the limit of large volume $R^3$. We calculate the condition this imposes on the power spectrum of a Gaussian random field. We quantify the bias for finite $R$, and show that the $R\to\infty$ limit is of little relevance for cosmological observations when the measured scales and correlations extend to the size of the observable universe. We consider curvature, density, and velocity perturbations. On large scales the bias is important in all three cases. For the density perturbations, which are observationally the most relevant, the relative bias first exceeds 100% at the separation $r=177$ Mpc, and is larger than 100% for all $r>560$ Mpc. It should be taken into account when comparing ensemble and volume averages for large-scale structure. The bias is also large for the cosmic microwave background temperature perturbations on large angular scales, but this is not relevant for observations, as their analysis does not involve volume averaging.
Read the full abstract
Metric Geometry
Approximate Gromov–Hausdorff continuity of magnitude and weighting
Magnitude gives an effective size of a finite metric space and is now used in several data-analysis settings. For such applications, it is natural to ask how magnitude behaves under Gromov-Hausdorff perturbations, including collisions of points. However, magnitude is nowhere continuous on finite metric spaces with the Gromov–Hausdorff topology. We show that this failure is nongeneric in a precise measure-theoretic sense. After fixing the number of points in each collapsing cluster, magnitude is approximately continuous. More strongly, when clusters of points collapse to the points of a limit space, the total weighting of each cluster approximately converges to the weighting of the corresponding limit point. The main tool is a general result on approximate limits of inverses near singular matrices.
Read the full abstract
Signal Processing
Channel Gains to Captions: Task-Unified Multi-Level RF Sensing with Vision-Language Models
This letter investigates a task-unified multi-level radio-frequency (RF) sensing framework driven by vision-language models (VLMs). Existing RF sensing methods rely on task-specific designs and provide only partial environmental information, limiting their ability to handle emerging 6G applications. To address this, we propose a generative formulation for RF sensing, where millimeter-wave (mmWave)/terahertz (THz) channel gains are mapped to captions describing multi-level environmental semantics. The framework solves this problem through a complementary design for RF-environment semantic bridging, where a VLM is fine-tuned to leverage its multimodal representations and prompt-conditioned semantic generation capabilities. Hence, different sensing tasks are specified through textual prompts, enabling the framework to handle diverse tasks in a unified manner. For fine-tuning, we introduce prompt-routed low-rank adaptation (LoRA) experts to achieve level-aware adaptation. Simulation results show that, compared with baselines, our framework achieves superior performance with a broader semantic scope, and enables task-unified sensing beyond predefined tasks. Under an unseen sensing requirement, it achieves an average F1-score improvement of 0.17 over the most competitive variant.
Read the full abstract
Computer Vision
Reliable Benchmarking of Artifact Detection in Computational Pathology: A Reproducibility and Uncertainty Analysis
Background and Objective: Quality control is a prerequisite for whole-slide image analysis, yet the benchmarks on which quality-control methods are compared share four properties that make their reported differences hard to interpret: few independent slides, annotation concentrated in a minority of them, pooled ratio metrics with no closed-form standard error, and a single inherited train/test partition. We propose a reliability protocol for such benchmarks. Methods: The protocol quantifies four sources of variability - test-set sampling, training stochasticity, partition composition, and undocumented preprocessing - a claim is reportable only if it survives all four; three of the four cost minutes of compute. We apply it to an independent reconstruction of a published diffusion-based artifact detector, evaluated on the original 24-slide partition and against a supervised baseline. Results: The method's central mechanism reproduces: the auxiliary contrastive term improves pooled F1 from 0.673 to 0.688 and replicates under a second seed (+0.0156, p = 0.031; +0.0190, p = 0.005), although it acts on pen marking rather than the artifact types cited to motivate it. Its comparative claims do not: differences between design variants, and against the supervised baseline, fall inside the uncertainty of the evaluation. Four of 24 slides carry 70% of scored annotated pixels, giving an effective sample size of 6.2, and the inherited partition sits at the 7th percentile. An unreported tissue-restriction step excludes 41.4% of out-of-focus annotation against 2.6% of air bubble; such a gate is confounded with blur by construction. Conclusions: Small-cohort benchmarks support far weaker conclusions than current reporting implies. The four checks are cheap enough to accompany any evaluation on such a resource and separate reproducible effects from differences the evaluation cannot resolve.
Read the full abstract
Quantum Physics
Metrological quantum-to-classical crossover in the volume of a noisy quasiperiodic lattice
Localization-delocalization transitions have recently been proposed for building a class of efficient quantum many-body critical sensors. This work scrutinizes metrological performances of such devices by focusing on the Aubry-André-Harper model that supports a localization-delocalization transition at finite strength of the onsite potential. We identify a metrological quantum-to-classical crossover driven by the interplay between noise and system size, whereby quantum-enhanced scaling of the quantum Fisher information persists only up to a finite, noise-dependent characteristic system-size. We first consider thermal noise and show that, at and near the localization-delocalization transition, the quantum Fisher information exhibits quantum-enhanced scaling for small systems but the system is stripped of this advantage beyond the characteristic crossover length. The crossover length decreases with increasing temperature. We then consider imperfections in the lattice hopping strengths and find a qualitatively similar crossover. There, the quantum-enhanced regime, identified with super-extensive scaling, gives way to an extensive scaling-a classical-limited weaker form-at sufficiently large system sizes. Thus, distinct noise mechanisms lead to a common limitation on the scalability of quantum-enhanced sensing: increasing the probe size beyond a noise-dependent limit can destroy the metrological quantum advantage.
Read the full abstract
Materials Science
Role of self-coherence in single-electron phase contrast imaging
The extension of coherent lattice contrast into the energy loss region in high-resolution transmission electron microscopy (HRTEM) is described by a pulse-like electron-sample interaction in the energy/time uncertainty limit. It generates a wave packet by electron self-interference in any coherent-inelastic scattering event with energy loss. The width of this wave packet is characterized by a self-coherence length ls({\Delta}E) that is predictable because an intrinsic decoherence phase around one radian is set by the expectation value for phase fluctuations. In this case the visibility of interference contrast from a crystalline sample with lattice parameter a is limited by a Rayleigh-like transfer factor P(ls, a) in the self-coherently illuminated sample area. The model is verified by energy-filtered HRTEM images of hexagonal BN and identifies energy-loss-induced phase noise as a single-electron visibility limit distinct from resolution limitations caused by ensemble-coherence or counting-statistical noise.
Read the full abstract
Robotics
A Dual-Cam Parallel Elastic Actuator with Shared Gas-Spring Compensation for Humanoid Ankles
To improve torque capacity and energy efficiency of humanoid ankles, this paper proposes a 2-DoF parallel elastic actuator (PEA). The main novelty of the proposed design lies in its dual-cam, single-gas-spring architecture, which enables torque compensation in both pitch and roll using a shared elastic element, thereby improving structural compactness compared with conventional multi-element compensation schemes. By leveraging parallel gas springs and customized cam modules, the proposed architecture provides dual-axis torque assistance tailored to specific task requirements. The second key contribution is the formulation of a coupled 2-DoF mathematical model that explicitly captures the interdependence between the two compensation units through the shared spring. Based on this model, an optimization-based design framework is developed to synthesize customized cam profiles from prescribed torque references, establishing a systematic link from task requirements to hardware realization. The complete lower-leg CAD integration is presented in detail. Static FEA and kinematic simulations confirm the design's feasibility and torque-relief effectiveness. The results highlight the proposed design as a compact, customizable solution for 2-DoF humanoid ankle torque compensation.
Read the full abstract
Functional Analysis
Sharp discrete Hardy constants in dimensions three and four and strict upper bounds from dimension nine
For $N\ge3$, let $C(N)$ be the optimal constant in the nearest-neighbour Hardy inequality on $\mathbb Z^N$ with $u(0)=0$, and set $A_N=(N-2)^2/4$. We prove that the continuum coefficient remains an upper bound, $C(N)\le A_N$, in every dimension, and determine the exact values $C(3)=A_3=1/4$ and $C(4)=A_4=1$. In higher dimensions we show $C(N)<A_N$ for $N=9,10$ and obtain the explicit bound \[ C(N)\le 3N-\sqrt{N^2+8N-8}<2N\qquad(N\ge3), \] which lies below $A_N$ from dimension eleven onward. The low-dimensional equalities follow from shifted radial supersolutions, angular convexity, and a discrete ground-state representation. We also show that this shifted-power mechanism cannot work at the continuum coefficient from dimension five onward. Dimension nine is treated by a Gaussian Rayleigh–Ritz construction combined with exact Jacobi theta-function estimates, while the higher-dimensional bounds are obtained through finite-dimensional orbit compressions. Finally, we derive positive spatial remainders in dimensions three and four and a spectral consequence for the associated discrete Schrödinger operators in the high-dimensional regime.
Read the full abstract
Operating Systems
Adaptive KV Retention for LLM Agents at Human-Approval Timescales
Unlike the seconds-scale tool-call pauses targeted by prior agent-serving systems, agentic LLM requests can be suspended for minutes or hours while waiting for human approval. We study how suspension and resumption affect GPU serving performance and develop a retention policy that balances active-serving capacity against future recomputation under uncertain approval waits. The central tension is severe because retaining suspended KV preserves fast resume but can consume enough GPU capacity to reduce active-serving goodput by 41%, while evicting it avoids that residency cost at the expense of nearly $10\times$ higher resume latency when the request returns. We develop a tiered retention controller around GPU opportunity cost, which expresses the serving capacity consumed by preserving or reconstructing a suspended request's KV state in a common GPU-time cost. Within host memory, the controller selects between indefinite retention and load-indexed expiration using calibration wait samples, without requiring per-request wait prediction. On human-scale approval workloads, our controller improves active-request goodput by 23-51% over the vLLM baselines, 22-29% over MORI, and 41-52% over Continuum.
Read the full abstract
Probability
Free energy and phase transition for 2D directed polymers with critical spatial correlations
We study the two-dimensional directed polymer model in a Gaussian environment which is independent in time and spatially correlated, with covariances $h(x)$ either summable or with a critical decay, satisfying $h(x) \sim (\log |x|)^a/|x|^2$ as $|x|\to\infty$ for some $a>-1$. We determine the precise high-temperature asymptotics of the free energy, confirming a conjecture of Lacoin (Ann. Probab. 2011), later refined by Cosco, Cottini and Donadini (2025). We also establish a phase transition for the diffusively rescaled partition functions: below some critical point they converge to the Lebesgue measure, while above it they converge to zero. A key feature of our approach is that both results are obtained using only second-moment estimates and are based on a simplified change-of-measure argument in the supercritical regime, that may prove useful for other disordered models.
Read the full abstract
Computation and Language
Opinionated, Hesitant and Stressed: Three Studies of How Politicians Speak in Four Slavic Parliaments
We present three large-scale studies of spoken parliamentary speech across four Slavic languages (Croatian, Czech, Polish, Serbian), drawing on over 6,000 hours from the ParlaSpeech 3.0 corpus. The first study examines how utterance-level sentiment shapes acoustic realisation: negative speech is consistently produced with higher pitch, greater intensity, and faster rate across all four parliaments, with a secondary arousal-driven upturn at the most positive extreme. The second study models filled pause frequency using negative binomial GEE, finding that speech rate, age, and sentiment are robust cross-lingual predictors, while gender effects reverse between South Slavic (men produce fewer filled pauses) and West Slavic parliaments (no gender difference) - a pattern invisible to single-language designs. The third study investigates primary stress variation in Croatian, showing that speaker-level preferences for early versus late stress cohere across verbs, adjectives, and nouns but decouple for adverbs and proper nouns. We conclude with a research agenda spanning corpus phonetics, disfluency modelling, and political rhetoric.
Read the full abstract
Computation and Language
Error-Type-Aware Loss Reweighting for Robust Named Entity Recognition with Noisy LLM Labels
Large language models are increasingly used to annotate datasets for training smaller, task-specialized models such as named entity recognition. While this method yields effective models, it assumes that the synthetic dataset is correctly annotated. In this work, we find that (i) current fine-tuning processes simply ignore LLM-introduced annotation noise, resulting in degraded performance and (ii) existing noise-robust losses are not transferable to sequence labeling because annotation noise in named entity recognition is heterogeneous: for example, missing mentions and type errors affect the training signal in different ways. Treating all noisy tokens equally in noise-robust losses and applying a single reweighing criterion for all may therefore remove useful supervision or reinforce incorrect labels. To address this limitation, we propose error-type-aware loss reweighting for NER, which introduces separate reweighing rules for different types of potentially erroneous tokens. Our approach is simple and efficient, does not require additional training resources, and improves F1 by 0.8 - 2.0 percentage points on dataset-level average for noise levels between 15% and 40%, with a maximum improvement of 4.6 percentage points with 24.1% noise on Wikigold.
Read the full abstract
Signal Processing
Explainable deformable matched filtering reveals measurable departures from classical receiver theory in optical wireless communications
Matched filtering is a central result of communication theory, providing the optimal linear receiver when the received waveform satisfies specific assumptions. Practical communication systems rarely satisfy these assumptions, yet learned receivers that outperform the classical matched filter provide little insight into what those improvements reveal about the limitations of the underlying theory. Here we introduce an explainable deformable matched-filter framework in which machine learning is constrained to learn a low-dimensional deformation of the classical matched filter rather than replacing it. Because every learned correction is defined relative to the theoretical matched-filter solution, the deformation becomes a measurable representation of receiver mismatch rather than an unconstrained optimisation. The communication waveform remains processed entirely by the matched filter, while a Kolmogorov-Arnold Network predicts only the deformation from physically interpretable receiver-state descriptors. Using an optical wireless communication testbed spanning ten signalling formats, four impairment classes and 1,600 conditions, we show that learned deformations improve receiver performance, yielding a median relative error-vector-magnitude reduction of 18.1%, while revealing departures from classical matched-filter optimality. Different signalling families occupy distinct deformation regimes, spectral analysis identifies the physical mechanisms underlying receiver mismatch, and latent receiver-state organisation demonstrates that these departures are structured rather than arbitrary.
Read the full abstract
Systems and Control
Real-Time Reference Shaping for Servo Systems
This paper addresses real-time motion planning for servo systems subject to nonlinear, state-dependent actuator constraints. A reference reshaping method is proposed that combines analytical optimality with high computational efficiency. Using Karush-Kuhn-Tucker conditions, the problem structure is explicitly characterized, and it is shown that the optimal solution lies within a finite set of candidate points. The complete solution set is constructed via closed-form expressions and a small-scale eigenvalue problem, yielding a deterministic algorithm that recovers the exact optimal solution without iterative optimization or root-finding. To address kinematic mismatch induced by aggressive commands, a real-time trajectory compensator is introduced to correct accumulated position error while preserving feasibility. Simulation results demonstrate significant computational speed improvements over existing methods, enabling real-time implementation at frequencies approaching 100 kHz.
Read the full abstract
Computer Vision
Whole-Body MRI Classification via Prompt-Based Clinical Conditioning
Combining whole-body magnetic resonance imaging (WB-MRI) with clinical variables has the potential to improve systemic disease diagnosis by leveraging complementary sources of patient information. However, structured clinical variables are often incomplete or missing, limiting the applicability of conventional multimodal fusion methods that assume fixed inputs. In this work, we propose TACTIC (Tabular-Attribute Conditioned Transformer for Image Classification), a prompt-based multimodal framework that integrates WB-MRI and structured clinical data through conditional visual feature learning. By encoding clinical attributes as prompts, TACTIC supports an arbitrary number of tabular inputs and naturally handles missing data without requiring imputation or fixed input structures. We evaluate TACTIC on five WB-MRI classification tasks spanning systemic and oncologic applications, including diabetes, chronic obstructive pulmonary disease (COPD), breast cancer, prostate cancer, and metastasis diagnosis. Across all tasks, TACTIC consistently improves performance over image-only baselines when clinical information is available while maintaining strong predictive capability under incomplete tabular inputs. Our results demonstrate the effectiveness of prompt-based models as a flexible approach for improving WB-MRI analysis using clinical context. The model weights and code are available at https://github.com/lauradaza/TACTIC
Read the full abstract
Sound
Vocal Music under Phoneme-Conditional Analysis
The vocal music of each language carries a distinctive sonic identity, even without instrumental accompaniment. We ask whether these differences are measurable and traceable to specific phonemes. To tackle this question, we introduce phoneme-conditional analysis, which isolates the acoustic effect of typologically distinctive phonemes by comparing marker syllables against matched non-marker controls within the same song, holding singer, melody, and genre constant. Across nine typologically diverse languages and thousands of songs, we measure effects along five acoustic dimensions. Song-level profiles built from these effects identify the language of an unaccompanied vocal at 85.5% balanced accuracy in a nine-way classification with folds grouped by artist; whether the separability arises by accumulation of the phoneme-local effects themselves is left open. Our findings suggest that phonological structure leaves systematic and measurable traces in how each language is sung.
Read the full abstract
Optics
Site-Selective Activation of Near-Infrared Emitters in hBN
Near-infrared (NIR) emitters in hexagonal boron nitride (hBN) have recently emerged as bright and spectrally stable light sources, but they have so far been found at locations set by the local defect environment rather than by design. Here we show that NIR emitters can be activated at predefined sites by combining single-shot femtosecond laser writing with annealing in an oxygen-rich environment. To our knowledge, this is the first demonstration of site-selective activation of NIR emitters in hBN. The written matrices are recovered in post-anneal photoluminescence maps, including in collection bands centred at 800 and 900 nm. Site-resolved spectra under 532, 635 and 785 nm excitation show narrow lines extending to 1 um; 83% of the spectra acquired at written sites under 785 nm excitation and 71% under 635 nm excitation contain at least one resolved peak above 810 nm. The integrated emission increases with writing-pulse energy over the range studied, and many sites host more than one emitter, which identifies lower writing energies as the route to single-emitter occupancy. Photon-correlation measurements on selected sites show antibunching, with a raw, uncorrected second-order correlation value g(2)(0) of 0.31 in a band centred at 850 nm. Spectral time series reveal bands that persist over the recorded intervals together with intermittent emission beyond 900 nm. Comparison with identically written, argon-annealed samples indicates that the annealing atmosphere controls the spectral range of the activated emitters, while the laser defines where they form.